Further contributions to smart grids cyber-physical security as a malicious data attack: Proof and properties of the parameter error spreading out to the measurements and a relaxed correction model
Further contributions to smart grids cyber-physical security as a malicious data attack: Proof and properties of the parameter error spreading out to the measurements and a relaxed correction model
复制标题
DOI：
10.1016/j.ijepes.2018.06.039
复制
发表时间：
2019
影响因子：
5.2
通讯作者：
A. Bretas;N. Bretas;B. E. Carvalho
中科院分区：
文献类型：
--
作者：
A. Bretas;N. Bretas;B. E. Carvalho
本文提出了作为一种恶意数据攻击对智能电网网络物理安全的进一步贡献。这些贡献是双重的。首先，给出了参数误差如何在具有误差的参数的测量函数上扩散的形式证明。然后，对于这种误差情况，以其归一化形式演示最大合成测量误差属性。其次，提出了一种智能电网网络物理恶意数据注入纠正方法。当前最先进的解决方案在假设测量或参数没有错误的情况下纠正同时攻击。然而，如果参数可能同时出错或反过来，如何纠正测量结果？本文针对这一问题提出了一种松弛模型策略。仅使用测量粗差分析框架同时处理和分析攻击。网络攻击检测是通过对归一化合成测量误差进行卡方(χ2)假设检验来进行的。合成误差用测量的新息指数(II)进行估计。通过最大归一化误差检验性质对网络攻击进行识别。考虑网络攻击类型，在松弛模型策略中使用组合归一化误差(CNE)对网络攻击进行修正。提出的解决方案适用于恶意测量和参数数据攻击。尽管如此，状态估计软件不需要重大改变。在IEEE 14节点和57节点系统上进行了验证。
This paper presents further contributions to smart grids cyber-physical security as a malicious data attack. The contributions are twofold. First, a formal proof of how parameter errors spread out on the measurement function having a parameter with error. The largest composed measurement error property, in its normalized form, is then demonstrated for this case of error. Second, a methodology for smart grid cyber-physical malicious data injection correction is presented. Current state of the art solutions corrects simultaneous attacks assuming measurements or parameters without error. However, how may one correct a measurement if the parameter might be simultaneously in error or the other way around? In this paper, a relaxed model strategy for such is presented. Attacks are processed simultaneously and analyzed using only the framework of measurement gross error analysis. Cyber-attack detection is made through a Chi-square (χ2) Hypothesis Testing (HT) applied to the normalized composed measurement error (CMEN). Composed errors are estimated with measurements’ innovation index (II). Cyber-attack identification is made through the largest normalized error test property. Cyber-attack correction is made considering cyber-attack type and using the composed normalized error (CNE) in a relaxed model strategy. The proposed solution works for malicious measurement and parameter data attacks. Still, the state estimation software does not need major changes. Validation is made on the IEEE 14-bus and 57-bus systems.